Christian-
What is the future of the existing GeoGrid classes? Are these going to
be deprecated/removed or will they stick around for a while until the
GridCoverage API is well tested and stable? Since it is not backward
compatible, will you be providing documentation for client developers on
migrating from GeoGrid to GridCoverage?
Don
On 7/30/15 10:12 AM, Christian Ward-Garrison wrote:
Hi Antonio,
Yeah, our versions are a little convoluted right now. The current
release versions is 4.6.2, which is probably what you're using. We
also have two unreleased in-development versions: 4.6.3 and 5.0.0. We
decided to put GridCoverage (which replaces GeoGrid and fixes the
dateline issue) on 5.0.0 because it is a large change that isn't
backwards-compatible. Clients will have to rewrite parts of their code
to take advantage of that, but it couldn't really be helped. Better to
wrap those changes in a MAJOR-version increase (i.e. v4 -> v5) than a
PATCH-version increase (i.e. v4.6.2 -> v4.6.3) [1].
Cheers,
Christian
[1] http://semver.org/
On Thu, Jul 30, 2015 at 5:50 AM, Antonio Rodriges
<antonio.rrz@xxxxxxxxx <mailto:antonio.rrz@xxxxxxxxx>> wrote:
Christian:
Thank you.
Why NetCDF-Java v5.0? I used netcdfAll-4.6.jar [1], doesn't 4.6
reflect NetCDF-Java version?
[1]
http://www.unidata.ucar.edu/software/thredds/current/netcdf-java/documentation.htm
2015-07-30 1:48 GMT+03:00 Christian Ward-Garrison
<cwardgar@xxxxxxxx <mailto:cwardgar@xxxxxxxx>>:
> Hi Antonio,
>
> There are many longstanding issues concerning grids that cross
the date
> line, and this is one of them. It's actually a very hard problem
to solve in
> a way that doesn't break existing clients (e.g. the IDV), but we
finally
> have a solution for it in the upcoming NetCDF-Java v5.0. Until then,
> hopefully you can work around the date line weirdness.
>
> Cheers,
> Christian
>
> On Mon, Jul 27, 2015 at 6:27 AM, Antonio Rodriges
<antonio.rrz@xxxxxxxxx <mailto:antonio.rrz@xxxxxxxxx>>
> wrote:
>>
>> Hello,
>>
>> Am I doing smth wrong? -
>>
>> When I call GridCoordSystem.getRangesFromLatLonRect
>> I always get two-element List<Range> regardless of whether the
given
>> LatLonRect crosses dateline or not.
>>
>> There are some datasets with longitudes 0..360 and others with
>> -180..-180, so we have index breaks at 0 and +/-180
correspondingly.
>> When I call getRangesFromLatLonRect for box containing 0 it
returns me
>> the range for values > 0 and if crossing +180 it returns range
>> containing positive lons < 180.
>>
>> _______________________________________________
>> netcdf-java mailing list
>> netcdf-java@xxxxxxxxxxxxxxxx <mailto:netcdf-java@xxxxxxxxxxxxxxxx>
>> For list information or to unsubscribe, visit:
>> http://www.unidata.ucar.edu/mailing_lists/
>>
>
_______________________________________________
netcdf-java mailing list
netcdf-java@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
